3D Magnetotelluric Research Articles

Analysis for Geoelectric Strike Direction of Magnetotelluric Data from a Geothermal Area

Magnetotelluric (MT) method plays an important role in geothermal exploration. For MT data interpretation, 2D modelling is still performed in a routine basis since 3D MT modelling is computationally intensive and needs sigificant computational resources. In 2D modelling, information on geoelectric strike direction is a requirement. In this study, analysis of geoelectric strike was performed on MT data from a geothermal prospect using phase tensor and tipper strike. The results from phase tensor analysis and verified by tipper strike show that the geoelectric strike of the study area is about N300°E that agrees well with the regional structure of the study area. The MT tensor data were rotated to the dominant geoelectric strike and 2D inversion modelling was performed. The subsurface resistivity model reveals the geothermal system of the study area.

Geoelectric Strike Analysis from Magnetotelluric (MT) Data Using Swift and Polar Diagram Methods

The magnetotelluric (MT) method is commonly used to estimate the subsurface conductivity structure. From the point of view of computational resources availability, currently it is a common practice to perform a 2D MT inversion modelling of MT data measured along a profile. For that purpose, the geoelectric strike angle determination from MT data is essential. The geoelectric strike angle can be determined by rotational analysis of the MT impedance tensor. In this paper we compare geoelectrical strike angle from both Swift method and polar diagram analysis by plotting the results as rose diagrams. The results from rotational analysis suffer from 90° ambiguity in the structural direction. Along with information from the regional geology to resolve the ambiguity, the profile for MT 2D modelling can be more appropriately selected. We used MT data from a geothermal prospect to illustrate our exercise.

Geoelectrical structure of Ulleung Island, Korea: interpretation of three-dimensional magnetotelluric data

ABSTRACTA three-dimensional (3D) magnetotelluric (MT) survey was carried out to delineate subsurface structures and possible fractures for geothermal development around Ulleung Island, Korea. Good-quality MT data were obtained throughout the survey region with respect to a remote reference located in Sawauchi, Japan, ∼ 900 km from the centre of the field site. 3D modelling and inversion were then performed, considering the sea effect on MT measurements near the coastline. The modelling results showed that seawater surrounding Ulleung Island affected MT data at frequencies below 10 Hz, depending on the distance from the shore. Based on this information, two types of 3D MT data inversion were performed: ordinary 3D inversion and 3D inversion using sea effects as a constraint. 3D inversion using the sea-effect constraint gave more reasonable and reliable results. In addition, the inversion that accounted for sea effects indicated a clear east–west conductive anomaly along the line connecting test boreholes GH-1 and GH-2; a high geothermal gradient of ∼ 100 °C/km was recorded. The geothermal gradients were lower, < 80 °C/km, at boreholes located in the southern (GH-3) and northern (GH-4) regions of the study area. These results suggest that deep circulation of fluid through deeply connected fractures, e.g. the anomaly observed in the east–west direction, carries heat from the bottom depths to near-surface layers. This would explain the large differences in gradients among sites in the study area.

Particle swarm optimization of 2D magnetotelluric data

We implement the particle swarm optimization (PSO) algorithm for the two-dimensional (2D) magnetotelluric (MT) inverse problem. We first validate PSO on two synthetic models of different complexity and then apply it to an MT benchmark for real-field data, the COPROD2 data set (Canada). We pay particular attention to the selection of the PSO input parameters to properly address the complexity of the 2D MT inverse problem. We enhance the stability and convergence of the solution of the geophysical problem by applying the hierarchical PSO with time-varying acceleration coefficients (HPSO-TVAC). Moreover, we parallelize the code to reduce the computation time because PSO is a computationally demanding global search algorithm. The inverse problem was solved for the synthetic data both by giving a priori information at the beginning and by using a random initialization. The a priori information was given to a small number of particles as the initial position within the search space of solutions, so that the swarming behavior was only slightly influenced. We have demonstrated that there is no need for the a priori initialization to obtain robust 2D models because the results are largely comparable with the results from randomly initialized PSO. The optimization of the COPROD2 data set provides a resistivity model of the earth in line with results from previous interpretations. Our results suggest that the 2D MT inverse problem can be successfully addressed by means of computational swarm intelligence.

2D and 3D MT in the central Skellefte Ore District, northern Sweden

New broadband magnetotelluric (MT) data have been acquired along two parallel profiles in the central part of the metallogenic Skellefte district in northern Sweden. The data were recorded as part of the Swedish 4D modelling of mineral belts project and cover an area with several economical and sub-economical deposits. The dimensionality and quality of the data were carefully analyzed and new error floors were systematically determined prior to inverse modelling in 2D and 3D. The algorithms used were EMILIA and WSINV3DMT. For the 2D inversion, only the determinant of the impedance tensor was used, while for the 3D inversion all elements were considered. The obtained models fit the inverted data, and image the main regional features. A detailed comparison reveals the superiority of the 3D model, both in model structures and data fit. After assessing the main features in the model, an interpretation is proposed and refined with the support of previous geophysical studies. The most interesting features are large and medium-sized conductors associated with crustal-scale shear zones and faults within the Skellefte Group rocks. These may be depicting a network of fossil pathways for hydrothermal fluid transport and as such, provide new insight into past processes in the area.

Geothermal reservoir boundary delineation using 3D magnetotelluric inversion. case study: the “delta” geothermal field of Indonesia

The “DELTA” geothermal field is a brown geothermal field in Indonesia. The field has been produced for many years. Most of productions wells are in the center of the reservoir, need to have step-out wells to reduce the well intervention risk. The first step-out well was drilled near reservoir boundary of conceptual model which is refer to “old” 3D Magnetotelluric (MT) inversion. The step-out well failed to produce steam and well data indicated outside the reservoir. After careful analysis, it found that the 3D MT model does not match with the well data. On this study, the author conducts 3D MT inversion re-modeling to re-delineate the reservoir boundary. To have comprehensive interpretation, the result of new 3D model which generate independently then compared with available data sets such as MEQ, gravity, geology map, geochemistry data and well data. The new 3D MT model shows better match to well data, re-delineate the reservoir boundary and give recommendation for well targeting areas.

Inverting magnetotelluric responses in a three-dimensional earth using fast forward approximations based on artificial neural networks

The most computationally intensive step in 3D magnetotelluric (MT) inversion is the calculation of the forward response. This fact makes any modelling which requires many function evaluations, including genetic algorithms and Markov chain Monte Carlo inversion, extremely time consuming. Using Artificial Neural Networks (ANNs) it is possible to approximate these expensive forward functions with rapidly evaluated alternatives. Using a limited subset of resistivity models created in a simple parameterisation, this work is the first to apply ANNs to approximate the 3D MT forward function. Training data are generated with a compute time of two weeks, and after training the ANN is able to reproduce forward responses at arbitrary site locations with accuracy similar to the level of typical data errors. To evaluate the accuracy of the models, we show that these forward responses may be used to successfully invert MT data in an evolutionary framework. Examples are shown in both synthetic and real-world scenarios, and results are compared with those from traditional inversion algorithms. We conclude that the trained ANN inversion has a fraction of the run-time of a traditional inversion and is successful at modelling the space of its limited parameterisation.

Lithosphere structures of the Parnaíba Basin and adjacent provinces revealed by deep magnetotelluric imaging

Magnetotelluric (MT) data from 49 broadband and 23 long period stations, covering periods between 10−3 and 104 s, were measured along a NNW-SSE profile of approximately 470 km, crossing the Parnaíba Basin, Borborema Province and São Francisco Craton in northeastern Brazil. The geotectonic scenario of the study region comprehends a complex structural framework formed during the Brasiliano orogenic collage. The MT data were processed to obtain the impedance tensor estimates for each MT site using a robust technique and then followed by a dimensional analysis via both the rotational invariants (WAL invariants) of the magnetotelluric tensor and the phase tensor. The two approaches indicated a three-dimensional nature of the region, allowing to proceed to 3D inversion, despite the data distribution along a profile. Several inversion trials of the full impedance tensor, tipper and tipper combined with impedance tensor resulted in a final 3D resistivity model, which suggests a division of the profile into three major zones. A sub horizontal resistive crustal block, interpreted as being part of the Parnaíba block, characterizes the western zone. A mosaic of resistive and conductive zones, which integrates the extended fold belt of the Borborema Province, composes the central zone. In the interface between the central and the eastern zones, there is a sub vertical conductive region limiting two resistive blocks, interpreted as a suture zone. This suture divides tectonic provinces and is supposed to be related with a Neoproterozoic collisional system. The 3D MT imaging recovered the structural framework under the Parnaiba Basin and marked the main lineaments. The paper suggests a distinct tectonic evolution for the complex assemblage of blocks in Northeast Brazil.

3-D magnetotelluric inversion of 2-D profiles: Application to land and marine data for shallow crustal investigation

The 3-D magnetotelluric (MT) inversion has a great potential for providing useful information on the geological background needed in petroleum and mining exploration. We developed a simple and robust full tensor 3D MT data inversion scheme that was tested on several real data sets. The technique allows to provide a detailed 3-D image of complicated geological contexts such as rift basins beneath thick basaltic screen covers. However, for practical reasons, 2-D profiles are still carried out to investigate geological questions. In such cases, tensor decomposition helps to obtain a local or regional strike which enable 2-D inversion but static distortion and effects of local and regional 3-D structures may affect the data. Ad hoc static shift techniques and heavy smoothing techniques are often used to obtain a reasonable solution but they may induce possible bias in the result. Here, we propose a 3-D inversion approach to interpret 2-D data without any decomposition technique or static correction required. The 3-D inversion uses the full MT tensor information in order to recover structures on either side of the 2-D profile. We present results from a synthetic test and applications to land and marine real data sets. Comparison between a standard 2-D approach and our local 3-D inversion provides clues about the bias induced by 2-D inversion of 3-D structures.

Finite Difference Algorithm on Non-Uniform Meshes for Modeling 2D Magnetotelluric Responses

A finite-difference approach with non-uniform meshes was presented for simulating magnetotelluric responses in 2D structures. We presented the calculation formula of this scheme from the boundary value problem of electric field and magnetic field, and compared finite-difference solutions with finite-element numerical results and analytical solutions of a 1D model. First, a homogeneous half-space model was tested and the finite-difference approach can provide very good accuracy for 2D magnetotelluric modeling. Then we compared them to the analytical solutions for the two-layered geo-electric model; the relative errors of the apparent resistivity and the impedance phase were both increased when the frequency was increased. To conclude, we compare our finite-difference simulation results with COMMEMI 2D-0 model with the finite-element solutions. Both results are in close agreement to each other. These comparisons can confirm the validity and reliability of our finite-difference algorithm. Moreover, a future project will extend the 2D structures to 3D, where non-uniform meshes should perform especially well.

Exploring magnetotelluric model space

Magnetotelluric (MT) inversions are inherently non-unique. Due to the large computational requirement of 3D MT inversion, there is always a trade-off between exploring model space and the amount of time invested in the inversion process. A standard approach is a two-stage inversion, where coarse features are resolved first, and the outputs from the coarse model are used as the starting model for a finely discretized inversion. Inversion may be followed by a combination of sensitivity analysis and forward modelling to test the robustness of features. This approach quickly leads to a low RMS model but has a limited capacity to explore the potential range of acceptable models.In this paper, we propose a revised two-stage process which enables fast comparison of inversion results from a series of inversion parameters. Three inversions were run using a starting half space with only the input data varied: one inversion using un-rotated full tensor and tipper data; one inversion using only tipper data; and the third inversion using rotated tensor and tipper data. In addition to these half-space models, an inversion was run using starting models based on geological constraints. It used three domain boundaries with roughness penalties turned off between units. The modified two stage inversion process suggested in this paper offers a balance between arriving at final inversion within a reasonable time frame while still allowing the user to explore a larger range of possible models than the original methodology.

ECloud - Magnetotelluric Webapp

Geophysical web applications are highly underrated. There are many potential benefits for transitioning to a cloud based web applications including compatibility improvement, low on-going operating costs and better access to existing third-party code and tools for facilitating rapid development. For users, time spent on installation, general IT maintenance such license management and upgrading is minimized. While web applications for business style applications are deeply established, scientific applications on the web are only just emerging. This research builds upon our previous feasibility study (Pethick and Harris 2015), which was the development of a 1D MT web application. This is to be redevelop into a commercial grade cloud hosted geophysical inversion web app that can cooperatively invert seismic and magnetotelluric (MT) data. Our software, ECloud, is designed to be user friendly where geoscientists and drillers can upload MT field data directly from their laptop, tablet or mobile to obtain subsurface geo-electrical distributions quickly with minimal input. This application is designed to be scalable, suiting cloud environments and is currently hosted on an Amazon EC2 instance. The preconceived notion that web applications are slow will be challenged. The purposefully designed MT algorithm and software structure will hopefully result in lower computation times while minimizing restrictions based on hardware requirements (i.e., primary memory).

Comparative Analysis and Joint Inversion of MT and ZTEM Data

Ground Magnetotelluric (MT) data acquired today are typically broadband, covering 0.001 to >1000 Hz with inter-site spacing typically at 500 to 1000 m. Airborne Z-axis tipper data (ZTEM) are sampled at higher spatial density but usually band-limited to frequencies >30Hz. We analyze a pair of overlapping 3D surveys to examine lateral and vertical spatial sensitivity.The MT data include a 2D line and a 3D survey. The line data also has magnetic tipper data that allows for a direct comparison with ZTEM; in the overlapping frequency range the agreement between the two magnetic data sets is good, with ZTEM showing higher lateral smoothness.CGG’s RLM-3D non-linear conjugate gradient MT-CSEM inversion engine was extended to accurately model the ZTEM data, using measured sensor altimetry data and detailed 3D topography. Both single domain and joint inversions of the ZTEM and MT data were carried out. A suite of inversions were run to test the influence of starting resistivity and regularization parameters on output models, equally for MT, ZTEM, and joint MT+ZTEM inversions to allow for direct comparison.ZTEM single domain inversion results depend strongly on the starting resistivity value, confirming that the method maps relative variations rather than absolute resistivity values, as expected for magnetics-only measurements. Shallower lateral structure shows qualitative agreement with the MT, but at depth resistivity from ZTEM inversion is driven by model regularization only. Joint inversion improved the relatively shallow section, calibrating the ZTEM resistivities and adding continuity between the MT sites. Below around 1000m depth, the 3D resistivity model is controlled by the MT data alone. Our overall conclusion is that today’s 3D broadband MT only benefits from joint MT-ZTEM acquisition and inversion workflows in the case of sparse MT station spacing.

3D magnetotelluric inversions with unstructured finite-element and limited-memory quasi-Newton methods

Traditional 3D Magnetotelluric (MT) forward modeling and inversions are mostly based on structured meshes that have limited accuracy when modeling undulating surfaces and arbitrary structures. By contrast, unstructured-grid-based methods can model complex underground structures with high accuracy and overcome the defects of traditional methods, such as the high computational cost for improving model accuracy and the difficulty of inverting with topography. In this paper, we used the limited-memory quasi-Newton (L-BFGS) method with an unstructured finite-element grid to perform 3D MT inversions. This method avoids explicitly calculating Hessian matrices, which greatly reduces the memory requirements. After the first iteration, the approximate inverse Hessian matrix well approximates the true one, and the Newton step (set to 1) can meet the sufficient descent condition. Only one calculation of the objective function and its gradient are needed for each iteration, which greatly improves its computational efficiency. This approach is well-suited for large-scale 3D MT inversions. We have tested our algorithm on data with and without topography, and the results matched the real models well. We can recommend performing inversions based on an unstructured finite-element method and the L-BFGS method for situations with topography and complex underground structures.

Three-dimensional magnetotelluric imaging of the geothermal system beneath the Gonghe Basin, Northeast Tibetan Plateau

To explore the geothermal system in the Gonghe basin located between Kunlun and Qinlin Orogen in northeast Tibetan Plateau, 310 Magnetotelluric (MT) sites were deployed in the basin. The MT sites have an average spacing of 1∼2 km and almost cover the whole basin with an area of ∼100 km by 100 km. We first analyzed the MT data using phase tensor analysis to qualitatively estimate the dimensionality of the subsurface resistivity structure and the resistivity anomalies. We then determined a 3D resistivity model from the surface to the depth of ∼50 km through a 3D MT inversion of MT data at the 272 sites with relatively higher quality. Overall, the inverted resistivity model fits the MT data very well. Our 3D resistivity model reveals significantly conductive anomalies at depths above ∼3 km, which are interpreted as widespread hot dry rocks or hydrothermal reservoirs. Furthermore, the 3D resistivity model shows pronounced low resistivity zones in the middle crust (∼15-35 km). These anomalies trend a NW-SE direction, consistent with the Kunlun Fault strike. Between the near-surface conductive zone and the middle crust conductive zone, lie several vertical narrow conductive zones, which are generally associated with hidden faults identified in the basin. Synthetic tests are used to validate the resolution of large low resistivity anomalies in the middle crust and the vertically elongated low resistivity anomalies in the model. These faults may serve as the channels for transporting heat from the heat source in the middle crust to the shallow zone around the depth of 3 km. Our 3D MT imaging results characterize for the first time the 3D distribution of the geothermal system in the Gonghe Basin.

Modular implementation of magnetotelluric 2D forward modeling with general anisotropy

We present a general framework for two-dimensional finite difference modeling of magnetotelluric data in the presence of general anisotropy. Our approach is modular, allowing differential operators for a range of formulations of the governing equations, defined on several possible discrete grids, to be constructed from a basic set of first difference and averaging operators. We specifically consider two formulations of the two-dimensional anisotropic problem, one with Maxwell's equations reduced to a second order system in terms of three coupled electric components, and one in terms of coupled electric and magnetic x-components. Both formulations are discretized on a staggered grid; the second (coupled electric and magnetic) system is also implemented on a grid with fixed nodes (i.e., not staggered). The three implementations are validated and compared using a range of test models, including a half-space with general anisotropy, an infinite fault with axial anisotropy and a simple dyke model. Comparisons to analytic results (for half-space and fault models), and to results from other anisotropic codes, combined with grid-refinement convergence tests, demonstrate that our algorithms are accurate and capable of routine modeling of two-dimensional general anisotropy. These finite difference codes, demonstrating the flexibility of our numerical discretization approach, can be readily applied to other problems.

Probabilistic Magnetotelluric Inversion with Adaptive Regularisation Using the No-U-Turns Sampler

We present the first inversion of magnetotelluric (MT) data using a Hamiltonian Monte Carlo algorithm. The inversion of MT data is an underdetermined problem which leads to an ensemble of feasible models for a given dataset. A standard approach in MT inversion is to perform a deterministic search for the single solution which is maximally smooth for a given data-fit threshold. An alternative approach is to use Markov Chain Monte Carlo (MCMC) methods, which have been used in MT inversion to explore the entire solution space and produce a suite of likely models. This approach has the advantage of assigning confidence to resistivity models, leading to better geological interpretations. Recent advances in MCMC techniques include the No-U-Turns Sampler (NUTS), an efficient and rapidly converging method which is based on Hamiltonian Monte Carlo. We have implemented a 1D MT inversion which uses the NUTS algorithm. Our model includes a fixed number of layers of variable thickness and resistivity, as well as probabilistic smoothing constraints which allow sharp and smooth transitions. We present the results of a synthetic study and show the accuracy of the technique, as well as the fast convergence, independence of starting models, and sampling efficiency. Finally, we test our technique on MT data collected from a site in Boulia, Queensland, Australia to show its utility in geological interpretation and ability to provide probabilistic estimates of features such as depth to basement.

Three-dimensional magnetotelluric axial anisotropic forward modeling and inversion

Magnetotelluric (MT) data has been widely used to image underground electrical structural. However, when the significant axial resistivity anisotropy presents, how this influences three-dimensional MT data has not been resolved clearly yet. We here propose a scheme for three-dimensional modeling of MT data in presence of axial anisotropic resistivity, where the electromagnetic fields are decomposed into primary and secondary components. A 3D staggered-grid finite difference method is then used to resolve the resulting 3D governing equations. Numerical tests have completed to validate the correctness and accuracy of the present algorithm. A limited-memory Broyden–Fletcher–Goldfarb–Shanno method is then utilized to realize the 3D MT axial anisotropic inversion. The testing results show that, compared to the results of isotropic resistivity inversion, taking account the axial anisotropy can much improve the inverted results.

Resistivity Image from 2D Inversion of Magnetotelluric Data in the Northern Cascadia Subduction Zone (United States)

Cascadia Subduction Zone (CSZ) lies between Pacific margin and North America plate. The purpose of this research is to identify the CSZ along Oregon, Idaho, Wyoming from conductivity (σ) contrast in the subsurface by using the magnetotelluric (MT) method. MT is an electromagnetic method that use frequency between 10-4 Hz and 104 Hz. We obtained the MT data from the EarthScope USArray in the form of EDI-File (five components of the electromagnetic field). We analyzed the MT data using phase tensor and modeled the data using 2D inversion. From the phase tensor analysis, the 3D data dominated the eastern regions. Global data misfit is 6,88, where WYI18 (close to Yellowstone) contributes misfit of 29,3. This means that the model response does not fit the data, which implies the data is not fully 2D. The 2D inversion results are found high resistivity anomalies (more than 500 ohm.m) at shallow depth beneath Oregon and Wyoming, which coresspond to high density anomalies. This high resistivity anomalies might correspond to the north American plate. Thus, it can be concluded that 2D inversion model can be used for most 3D MT data to illustrate the resistivity distribution in the Cascadia Subduction Zone.

The Use of Sparse Direct Solver in Vector Finite Element Modeling for Calculating Two Dimensional (2-D) Magnetotelluric Responses in Transverse Electric (TE) Mode

The late research, linear matrices of vector finite element in two dimensional(2-D) magnetotelluric (MT) responses modeling was solved by non-sparse direct solver in TE mode. Nevertheless, there is some weakness which have to be improved especially accuracy in the low frequency (10-3 Hz-10-5 Hz) which is not achieved yet and high cost computation in dense mesh. In this work, the solver which is used is sparse direct solver instead of non-sparse direct solverto overcome the weaknesses of solving linear matrices of vector finite element metod using non-sparse direct solver. Sparse direct solver will be advantageous in solving linear matrices of vector finite element method because of the matrix properties which is symmetrical and sparse. The validation of sparse direct solver in solving linear matrices of vector finite element has been done for a homogen half-space model and vertical contact model by analytical solution. Thevalidation result of sparse direct solver in solving linear matrices of vector finite element shows that sparse direct solver is more stable than non-sparse direct solver in computing linear problem of vector finite element method especially in low frequency. In the end, the accuracy of 2D MT responses modelling in low frequency (10-3 Hz-10-5 Hz) has been reached out under the efficient allocation memory of array and less computational time consuming.